New Airtrol tank fitting

jesmed1

Today I installed a new Airtrol tank fitting in our plain steel compression tank, thanks to advice from @EdTheHeaterMan and @leonz. The old setup had 1/2" horizontal pipe with no pitch running from both boilers to the expansion tank. There was a very old brass fitting on the tank inlet with an air vent tube to set the water level, but it was pretty rudimentary. It had a very small hole for the water flow in/out of the tank, like 3/16" diameter. Maybe that was an early attempt at preventing gravity circulation.

Anyway, the new Airtrol is piped with 3/4" copper pitched as steeply as I could given the physical envelope. The Airtrol installation instructions said steeper pitch was better, with a minimum of 1" rise per 5 foot run. I had enough space to do a 3-in-12 pitch.

After installing the Airtrol and sweating the pipes, but before filling the tank, I put a compressor fitting on the expansion tank drain valve and ran some compressed air into it to soap bubble leak check my joints. The sweated joints were all good, but the threaded adapter into the expansion tank leaked, so I had to cut the copper pipe near the Airtrol fitting, unscrew the fitting from the tank, re-tape and dope the adapter, screw it back in, and sweat the pipe back together with a slip coupling. Fortunately that fixed the one leak, and I refilled the tank with water, using the Airtrol bleed screw to set the water level. With cold weather coming tonight and tomorrow, the boilers will be cycling again soon, and hopefully that helps solve our air problem.

I'm also going to put a sign on the tank saying DO NOT DRAIN!

The second pic shows the new 3/4" copper sloping up from right to left to the tank. The third and fourth show the old tank inlet fitting with the air vent tube and the small inlet hole for the water.

Thanks again for everyone's help.

EdTheHeaterMan

Today I installed a new Airtrol tank fitting in our plain steel compression tank, thanks to advice from @EdTheHeaterMan and @leonz. The old setup had 1/2" horizontal pipe with no pitch running from both boilers to the expansion tank.

Are you saying that both boilers get connected to one tank?

You really should have separate tanks for each boiler.

But since that has (almost) worked for years, I don see why you need to change it. If I knew that, Might have recommend that you get two #30 Extrols (or at least one)

EdTheHeaterMan

I refilled the tank with water, using the Airtrol bleed screw to set the water level.

. Not really the right thing to do. You don't use the "bleed fitting" to get the water up to that level. You use that fitting to get air into the tank when you drain it so it is empty at cold start. when you open the valve to let system pressure in to the tank, the water will find its level without letting any air out. You don't want to let any air out when the tank is under pressure.

jesmed1

EdTheHeaterMan said:

I refilled the tank with water, using the Airtrol bleed screw to set the water level.

. Not really the right thing to do. You don't use the "bleed fitting" to get the water up to that level. You use that fitting to get air into the tank when you drain it so it is empty at cold start. when you open the valve to let system pressure in to the tank, the water will find its level without letting any air out. You don't want to let any air out when the tank is under pressure.

Thanks, Ed. I just followed the Airtrol operating instructions, which say:

"1. Close all air vents except the vent at the bottom of the Airtrol tank fitting.
2. Begin filling the system with fluid.
3. When water runs freely from the Airtrol tank fitting, close the vent tightly."

That's from page 2 of the Airtrol instructions in the attached pdf. So apparently the Airtrol people want the air vented from the tank until the water level reaches the top of the air vent tube. At that point, the air vent gets closed and the water level continues to rise until pressure equilibrium is reached.

I understand that will result in more water and less air in the tank than if I had started filling the tank with the air vent closed. But I figured they wrote the instructions that way for a reason, so I did what they told me. Maybe they want a higher water level because it will result in a smaller water-to-air contact surface area, and that should reduce how much air dissolves into the water?

Out of curiosity, I'm going to wait for a boiler run today and then put a thermal imager on the expansion tank to see if I can see the level of the (warm) water in the tank. If it does look too high, I can redo the fill procedure differently to get a lower level.

jesmed1

EdTheHeaterMan said:

Today I installed a new Airtrol tank fitting in our plain steel compression tank, thanks to advice from @EdTheHeaterMan and @leonz. The old setup had 1/2" horizontal pipe with no pitch running from both boilers to the expansion tank.

Are you saying that both boilers get connected to one tank?

You really should have separate tanks for each boiler.

But since that has (almost) worked for years, I don see why you need to change it. If I knew that, Might have recommend that you get two #30 Extrols (or at least one)

Yes, both boilers piped to one expansion tank. With the new Airtrol and better piping, it can't be worse than it was, and hopefully is at least somewhat better. The Airtrol and the pipes and fittings cost us less than one new diaphragm tank, so it was a low-cost improvement that the other owners were willing to pay for. Anything more expensive will have to wait until we need new boilers.

Steamhead

That thing you took out was a tank drainer. You installed it with a standard boiler drain in the bottom and opened the plug on the side when it was time to drain the tank. If the pipe to the tank was connected to it, that was wrong.

jesmed1

Steamhead said:

That thing you took out was a tank drainer. You installed it with a standard boiler drain in the bottom and opened the plug on the side when it was time to drain the tank. If the pipe to the tank was connected to it, that was wrong.

Yes, the pipe to the tank was connected to that fitting. Then there was an old hose bib-type faucet on the other end of the tank. So the original setup on the tank was a plain hose bib at one end for draining, and the other end had the old fitting I showed in the pic with the air bleed tube, which was also acting as the inlet/outlet for expansion water from/to the boilers.

I guess that originally, when this setup was first plumbed, the boiler techs drained the expansion tank by opening the hose bib at one end, and removing the air bleed screw (the "plug" as you called it) from this drain fitting at the other end. That probably went on until a few years ago, when a new boiler tech who didn't realize there was an air bleed screw in that fitting told me that I needed to install a Drain-O-Tank valve in the place of the hose bib, because the Drain-O-Tank has an air bleed feature that he thought was missing. So I got the Drain-O-Tank valve and installed it in place of the hose bib.

Then a year or two later, a more experienced tech who saw the new Drain-O-Tank valve told me that I already had an air bleed screw on the old fitting at the other end of the tank (which I had not realized until then), so the Drain-O-Tank wasn't necessary because it just duplicated an already-existing air bleed feature.

Anyway, now we have a Drain-O-Tank at one end, and the Airtrol at the other end. Hopefully that works better than the original setup.

Thanks for letting me know there was yet another thing wrong with our original setup... which makes me feel better about making the effort to install the Airtrol.

jesmed1

@EdTheHeaterMan Following up on the water level question, this morning both boilers ran simultaneously, so at the end of the runs, both water temps were at a maximum, which should represent maximum expansion. The pipe leading to the Airtrol was hot, but the tank surface felt cool, so however much hot water expanded into the tank wasn't enough to warm it noticeably.

So I took a heat gun and warmed the end of the tank evenly, being careful to distribute the heat uniformly across the entire area. I then used a thermal imager to take a number of photos over a span of several minutes. In all of them there was a clear delineation between the lower area that was being cooled by the tank water, which was around 75 degrees, and the upper area that wasn't in contact with water and thus retained more heat from the heat gun. (The red horizontal line across the bottom is a hot water pipe that was in the way.)

I then drew a line on the tank marking the upper boundary of the blue color, which I think marks the water level. (If I look closely, there's a very straight horizontal line where the blue color stops and the green starts above it.) The line is 9 inches up from the bottom, in a tank whose diameter is 14 inches. That works out to about 27 gallons of water and 13 gallons of air, in a 40 gallon tank. Concincidentally or not, that works out to almost exactly 2/3 water and 1/3 air. And this is at maximum water expansion. So I think our water level is OK.

I'm going to try this again later when the water has cooled off, and see if there's a noticeable difference in water level.


Update: After allowing the system to cool, I did the same thermal imaging process. This time the water level appears to be 7.5 inches, or 1.5 inches lower than at full expansion. This corresponds to a water volume of 22 gallons and an air volume of 18 gallons. So this is apparently the level that results from following the Airtrol instructions for filling, leaving the air bleed open until it begins to stream water, then closing the air bleed.

This is a slightly lower level than I would have expected. One reason may be that I had to use two adapter fittings to go from the 1/2" FPT Airtrol to the 3/4" FPT tank bung, and the height of the adapters pulled the Airtrol down about 1 inch lower, so the top end of the air vent tube was lowered by the same amount.

Anyway, the difference between the hot water volume and cold water volume in the tanks is 27-22=5 gallons, again based on the thermal images. While I was waiting for the system to cool off, and before I took the second image, I calculated our expected expansion volume using water starting at 60 F, ending at 140 F max temp, 300 gallons total for both boilers, and got an expected expansion volume of 5 gallons. I suspected that number was high, because I was generous in calculating water volume, and because the actual average water temperature delta is probably less, and I also neglected pipe expansion. But as chance would have it, the results of the thermal imaging (though admittedly imprecise) agreed exactly with the prediction of 5 gallons additional water volume by thermal expansion. However, as I said, the exact change in tank water volume is probably less, maybe more like 3 gallons.

leonz

Bravo!!!! You have the correct ratio of air blanket over water volume. now the heat will come easily.
As I write, my coal stoker boiler is dumping heat through the heating loop to drive out the dampness.

neilc

this may have been covered already,
and I'll bet I mentioned it in the other thread,
but, in the diagram, you're pumping towards the boiler and tank,
really should get those cirs on the supply side of the boulers,

jesmed1

neilc said:

this may have been covered already,
and I'll bet I mentioned it in the other thread,
but, in the diagram, you're pumping towards the boiler and tank,
really should get those cirs on the supply side of the boulers,

Yes, thank you. That's one of the improvements we'll make when a boiler needs replacing. The other will be to switch over to diaphragm tanks with air eliminators.